primitive data types - risc.jku.at · pdf file2.718281828459 2.7182817 2 wolfgang schreiner...

Primitive Data Types

Primitive Data TypesWolfgang Schreiner

Research Institute for Symbolic Computation (RISC)

Johannes Kepler University, Linz, Austria

[email protected]

http://www.risc.jku.at

Wolfgang Schreiner RISC



Each variable/constant has an associated data type.

• Primitive data types are built into the Java Virtual Machine.

boolean, byte, short, int, long, char, float, double.

•Other data types are implemented by programmers.

String.

A data type consists of a set of values and operations on these values.

Wolfgang Schreiner 1


Booleans

Data type boolean.

• Two constants:– true and false.

• Program:

boolean b = true;

System.out.println(b);

•Output:

true

The type of logical values.



Boolean Operations

Logical connectives.

Operator Description

! “not” (negation)

&& “and” (conjunction)

|| “or” (disjunction)

• Example:

System.out.println(!true);

System.out.println(true && false);

System.out.println(true || false);

false

false

true



Boolean Operations

• Precedence rules:– ! binds stronger than &&.

– && binds stronger than ||.

– ((!a) && b) || c can be written as

!a && b || c

• Short-circuited evaluation:– (false && any ) is false.

– (true || any ) is true.

– Second argument is not evaluated, if result already determined by first one.

Short-circuited evaluation will become important later.



Equality and Inequality

Relational operations defined on any data type:


== equal to

!= not equal to

• Program:

boolean b = (1 == 0);


•Output:

false

•Do we need the inequality operator?

We can decide about equality/inequality for any data type.Wolfgang Schreiner 5


Integers

Data types byte, short, int, long.

• Integral numerical values in a certain range:

Type Size Minimum Value Maximum Value

byte 8 bits -128 127

short 16 bits -32768 32767

int 32 bits -2147483648 2147483647

long 64 bits -9223372036854775808 9223372036854775807

• Predefined constants:Type Minimum Value Maximum Value

byte Byte.MIN VALUE Byte.MAX VALUE

short Short.MIN VALUE Short.MAX VALUE

int Integer.MIN VALUE Integer.MAX VALUE

long Long.MIN VALUE Long.MAX VALUE



Arithmetic Operations

• Literals:– 0, 9, 47113414.

– Larger than Integer.MAX VALUE: 49203281234L.

•Operations:


+ unary plus and addition

- unary minus and subtraction

* multiplication

/ truncated division

% remainder

• Truncated division: −17/2 = −8.– Remainder: a = (a/b) ∗ b + (a%b).

– What is −17%2?



Arithmetic Operations

• Example:

int a = 2; int b = 3;

a = ((a+b)*7)/a;

System.out.println(a);

17

•Division by zero:

int a = 1/0;

Exception in thread "main" java.lang.ArithmeticException: / by zero

at Main.main(Main.java:5)

• Usual precedence rules:– a*b+a/b

– (a*b)+(a/b)



Overflow

Integer types only represent subset of Z.

• Sum of two positive int values is not necessarily positive:

int a = 1500000000;

int b = 1000000000;

int c = a+b;

System.out.println(c);

-1794967296

• Integer overflow has occured:(1500000000 + 1000000000)−MAX VALUE + (MIN VALUE− 1)

= 2500000000− 2147483647 + (−2147483648− 1)

= 352516353− 2147483649

= −1794967296.

Programmer must ensure that this does not happen.Wolfgang Schreiner 9


Widening Conversions

• In arithmetic, byte and short values are converted to int:

byte a = 100;

byte b = 100;

System.out.println(a+b);

200

•Mixing of different integer types possible:

byte a = 127;

short b = 255;

int c = a*b+1;


32386

byte and short are promoted to int.Wolfgang Schreiner 10


Widening Conversions

• In arithmetic with long, int values are converted to long:

int a = Integer.MAX_VALUE;

long b = Integer.MAX_VALUE;

System.out.println(a+b);

4294967294

• In an assignment, the smaller type is promoted to the larger type:

byte a = 127;

short b = a;

int c = b;

long d = c;

System.out.println(d);

127

Widening conversions are automatically performed.Wolfgang Schreiner 11


Narrowing Conversions

• Narrowing conversions are not automatically performed:

int a = 127;

short b = a;

Main.java:6: possible loss of precision

found : int

required: short

• Problem with widening conversions:

short a = 1;

short b = a+1;

Main.java:6: possible loss of precision

found : int

required: short



Narrowing Conversions

• A type cast is required to convert to a narrower dataytype.– (type )expression

– value of expression is converted to type.

int a = 127;

short b = (short)a;

• Type cast binds stronger than arithmetic operations:

short a = 1;

short b = (short)(a+1);

– (short)a+1 would be useless.

Narrowing conversions require an explicit type cast.



Relational Operations


< less than

<= less than or equal to

> greater than

>= greater than or equal to

• Arithmetic promotion is applied:

short i = -1;

int j = 1;

boolean b = (i <= j);


true

•Does the following work?

boolean b = (-2 <= 1) || (1/0 == 1);



Characters

Data type char.

• The Unicode characters.– About 216 characters, two bytes of memory.

• Character literals: ’char ’– Program:

char letter = ’a’;

char newline = ’\n’;char backslash = ’\\’;System.out.print(letter);

System.out.print(newline);

System.out.println(backslash);

– Output:

a

\Wolfgang Schreiner 15


Special Characters

• Already known escape sequences:– \", \\, \n, \t

• Further escape sequences:– Single quote: \’– Hexadecimal Unicode value: \unnnn– Program:

char letter = ’\u0041’;System.out.println(letter);

– Output:

A



Characters as Integers

char is an integer data type smaller than int.

• Number interpreted as Unicode value.char letter = 65;

System.out.println(letter);

A

• Usage like any other integer type (e.g., short).boolean b = (’A’ <= ’B’)


true

• Promotion and type cast:int i = ’A’+1;

System.out.println((char)i);

B



Ordering Properties

• Upper-case letters are densely ordered:– ’A’, ’B’, . . . , ’Z’.

• Lower-case letters are densely ordered:– ’a’, ’b’, . . . , ’z’.

•Decimal digits are densely ordered:– ’0’, ’1’, . . . , ’9’.

• Example:– How to determine whether a character is a letter?

Look up Unicode table to find ordering properties.



Floating Point Numbers

Data types float and double.

• Fractional numerical values: −1.5, 0.0, 3.1415927.– Only approximated (no exact representation).

– Rounding errors may occur in computations: 1.5 + 1.5 6= 3.0

• Floating point:– No fixed number of digits after decimal point.

– Values close to zero are approximated with higher accuracy.

Type Size Minimum Value Maximum Value Significant digits

float 32 bits ca. −3.4 ∗ 1038 ca. 3.4 ∗ 1038 7

double 64 bits ca. −1.7 ∗ 10308 ca. 3.4 ∗ 10308 15

double represents larger values and with greater accuracy.



Floating Point Literals

Literals: at least one digit after the decimal point.

• By default, literals are of type double.

double g = 2.718281828459;

• Append F, to denote literal of type float.

float f = 3.1415927F;



Floating Point Operations

• Printing:

float f = 1.5F;

System.out.println(f);

1.5

• Arithmetic:Operator Description

+ unary plus and addition

- unary minus and subtraction

* multiplication

/ division

% remainder

a = (int)(a/b) ∗ b + (a%b).



Special Floating Point Values

• “Infinity”– Result of a computation is to big:

double f = 2.718281828459E300;

double g = f*f;

System.out.println(g);

Infinity

– Division by zero:

System.out.println(-1.0/0.0)

-Infinity

• “Not a Number”

System.out.println(0.0/0.0);

NaN



Conversions

• float is promoted to double.– Mixed arithmetic and assignments.

• Integer values are promoted to float values:float f = 1.5F;

long l = 1;

System.out.println(f+l);

2.5

• Narrowing conversion requires type cast:double f = 2.718281828459;

float g = (float)f;

int h = (int)f;

2.718281828459

2.7182817

2



Pitfalls

Floating point arithmetic is trickly.

float f = 3.141592F;

float g = 3.141593F;

System.out.println(f*f*f*f*f*f*f*f*f*f*f*f*f*g);

System.out.println(g*f*f*f*f*f*f*f*f*f*f*f*f*f);

9122149.0

9122147.0

• Associativity does not hold (a ∗ (b ∗ c) 6= (a ∗ b) ∗ c).

• Small rounding errors can accumulate to large overall errors.

Use floating points only when you need them and use them with care.



Relational Operations


< less than

<= less than or equal to

> greater than

<= greater than or equal to

• Example:int i = -1;

double d = 1.5;

boolean b = (i <= d);


true

• Exact comparisons do not make sense:– Wrong: f == g

– Better: Math.abs(f-g) < e



Strings

Sequences of characters represented by Unicode numbers.

String letter = "a";

String empty = "";

String word = "hello";

String sentence = "The ball is red.";

System.out.println(letter);

System.out.println(empty);

System.out.println(word);

System.out.println(sentence);

Output:a

hello

The ball is red.



String Literals

• String literals are delimited by double quotes.• The double quote has to be escaped in the literal.

String doubleQuote = "\"";System.out.println(doubleQuote);

Output:

"

While character literals are delimited by single quotes, string literalsare delimited by double quotes.



Unicode Codes

Unicode values may be used to denote characters in literals:

String letter = "\u0041";System.out.println(letter);

Output:A

Useful to denote characters that are not available on keyboard.



String Operations

• String concatenation: s1+s2

String s1 = "abc"; String s2 = "def";

String s = s1+s2;

System.out.println(s);

Output:

abcdef

• String length: s.length()

String s = "abc";

int n = s.length();

System.out.println(n);

Output:

3



String Operations

• Character extraction: s.charAt(i)

String s = "abcd"; char ch = s.charAt(2);

System.out.println(ch);

Output:

c

• Unicode value extraction: s.codePointAt(i)

String s = "abcd"; int c = s.codePointAt(2);


Output:

99

Index i must be less than s.length().



String Operations

• Substring extraction: s.substring(i,j) and s.substring(i)

String s = "abcdefg";

String t1 = s.substring(2,5);

String t2 = s.substring(2);

System.out.println(t1);

System.out.println(t2);

Output:

cde

cdefg

Index i denotes the position of the first character of the substring;index j denotes the position of the first character that is not in thesubstring any more.



String Operations

• String search: s1.indexOf(s2) and s1.lastIndexOf(s2)String s = "abcabc";

System.out.println(s.indexOf("bc") + " " + s.lastIndexOf("bc"));

Output:

1 4

• Generalization: s1.indexOf(s2,i) & s1.lastIndexOf(s2,i)String s = "abcabc";

System.out.println(s.indexOf("bc", 2) + " " + s.lastIndexOf("bc", 3));

Output:

4 1

The optional index i denotes the position where the search starts.



String Comparison

•Do not compare strings by == and !=

– "abcdef" == "abc" + "def" may not be true.

• String comparison: s1.equals(s2)– Returns true if s1 and s2 have the same length and the same character in every position.

String s = "abcdef";

System.out.println(s.equals("abc" + "def"));

Output:

true

Use equals to compare strings!



Lexicographical String Comparison

• s1.compareTo(s2) and s1.compareToIgnoreCase(s2)– Result is zero, if s1 and s2 are equal, negative, if s1 is before s2, positive, if s1 is after s2.

– First difference from the left determines the order between s1 and s2.

String s = "AB";

System.out.println(s.compareto("B") < 0);

System.out.println(s.compareTo("AC") < 0);

System.out.println(s.compareTo("ABC") < 0);

Output:

true

true

true

The well-known “phone book ordering”.



Basic Input

No simple input for basic data types in Java.

•Own input methods: kwm.jar (Input.README)

Method Description

Input.readBoolean() reads boolean value

Input.readByte() reads byte value

Input.readShort() reads short value

Input.readInt() reads int value

Input.readLong() reads long value

Input.readChar() reads char value

Input.readFloat() reads float value

Input.readDouble() reads double value

Input.readString() reads String value

• Reads one line from standard input.– Converts it to the specified data and returns the result.



Input Example

System.out.print("Enter first float: ");

float f1 = Input.readFloat();

System.out.print("Enter second float: ");

float f2 = Input.readFloat();

System.out.print("The product of " + f1 + " and " + f2 + " is ");

System.out.println(f1*f2);

Enter first float: 2.7

Enter second float: -1.4

The product of 2.7 and -1.4 is -3.78

What happens, if the input has wrong format?



Error Handling

• Control methods:Method Description

Input.isOkay() was last input okay?

Input.hasEnded() has input stream ended?

Input.getError() get message of last error

Input.clearError() clear error flag and message

• Check after input, whether result is okay.– If yes, use the returned value.

– If not, input has either ended or error has occurred.

– If error, get error message and clear error.

Details will be shown later.


primitive data types - risc.jku.at · pdf file2.718281828459 2.7182817 2 wolfgang schreiner...

Documents